Estimation of Soil Erosion and Sediment Transport by SWAT model (Case Study: Upstream of Badavar Basin, Lorestan)

The Soil and Water Assessment Tool – SWAT (Arnold et al. 1998; Arnold et al., 2012) is a well-established physically based, semi-distributed hydrological model for river basin scale application (Brighenti et al., 2019). is a hydrological model to assess sediment (Yesuf et al., 2015; Vigiak et al., 2017; Abdelwahab et al., 2018), impact of land use (Zeiger and Hubbart, 2016; Choto and Fetene, 2019; stream flow generation (Hallouz et al., 2018; in-stream water quality, climate change and, water quality and quantity variation (Jiao et al., 2014; Francesconi et al., 2016; Yang et al., 2018 ). A number of studies used laboratory, field scales and modeling studies to understand soil erosion and sediment dynamics in various regions. Several worldwide studies have been done using the SWAT model. The objective of this study is to apply the Soil and Water Assessment Tool (SWAT) model to predict surface runoff generation patterns and soil erosion hazard in Badavar catchment and to prioritize most degraded sub-catchment in order to adopt the appropriate management intervention. The specific objective of the study is validation of the SWAT model to assess its capability in predicting sediment yield transport in Badavar catchment.

Study area The study area is Badavar catchment (58311 ha) in the west of Iran and the northern part of Lorestan Province. Badavar catchment is one of the sub-catchments of of the Karkheh River. The average height of the catchment is 1941 m. The annual average precipitation is 527 mm, and the annual average temperature is about 10.5 °C. The climate of study area is in the category of sub-humid, characterized by cold winters and temperate summers. Most of the study area has slope 0-5 degree. Rain farming and week rangeland are the dominate land use in this area.The soil erosion is critical in the Badavar catchment. soil erosion in addition to damages to the land, it will reduce the capacity of the downstream dam of this catchment (the Karkheh dam). So in this study, SWAT model is used to identify the risk of erosion in Badavar catchment and various factors affecting of the erosion.SWAT Model: Model parameters: For preparation of input data, the GIS "version10.3"and ArcMap "ArcSWAT2012" was used. The following basic data were selected as SWAT model inputs:1) Digital Elevation Model (DEM): it is extracted from 1:50000 topography map with a spatial resolution of 30 m. 2) Landuse: it was obtained by Forests, Range and Watershed Management Organization of Iran. 3) Slope: it is extracted from DEM. Slope map shows that the Badavar catchment consists mainly of the plain with low slope (0-5 degree). 4) Hydro-meteorological data: The data required includes rainfall, river discharge, and climate data such as temperatures, solar radiations, humidity, and wind speed. These hydrological and meteorological data are collected from two organizations; the Iran Meteorological Organization and the Ministry of Energy.5) Soil data: SWAT model requires different soil textural and physico-chemical properties. These include soil texture, available water content, hydraulic conductivity, bulk density, and organic carbon content for different layers of each soil type. All of these soil characteristics and the soil map of the catchment were prepared by field survey and 39 systematic random samples of soil and laboratory works. Also the soil texture data was used to extract hydrological soil groups that were linked with FAO’s texture classification. This was then linked with the SWAT database using the soil layers and soil type. 6) Delineation of sub-catchments and HRUs:SWAT uses two types of functional units: the subbasin and the Hydrologic Response Unit (HRU) (Neitsch et al., 2011). The sub-basin is a spatially defined region that comprises a main reach and its contributing area, which is composed by one or more HRUs (Vigiak et al., 2017). On the other hand, analysis the catchment is allowed by SWAT as a whole or by subdividing it into sub-basins containing the same portions called Hydrological Response Units (HRU) (Briak et al., 2016). The HRU is a land unit of homogeneous environmental properties (soil, land use/cover, management, and topography) and hydrologic behavior (Vigiak et al., 2017). 33 sub-catchments and 53 Hydrological Response Units (HRU) have been generated for Badavar catchment.After simulation, factor analysis was used to determine the most important factors in sediment production. Factor analysis attempts to explain the correlations between the observations in terms of the underlying factors, which are not directly observable. The purpose of factor analysis is to reduce the complexity within the similarity matrix of a multivariate data collection, transforming it into a simpler and more easily interpreted factor matrix.Finally, for comparisons between observed and simulated sediment loads, four model evaluation statistics were selected; Correlation coefficient (R2), Root Mean Squared Error (RMSE), Index of Agreement (D) and Correlation determination (r).

The most important inputs of the model are: soil information, land use, slope, elevation, geology, weather information (rainfall, background and temperature, relative humidity, dew point, solar radiation and wind speed). Also factor analysis was used to determine the most important factors in sediment production. The simulation results showed that amount of sediment output from the basin is 7170 tons per year. After implementation of the model, the amount of sediment simulated and observation sediment was compared. Correlation coefficient (R2), Root Mean Squared Error (RMSE), Index of Agreement (D) and Correlation determination (r ) were validated with 95%, 03%, 97% and 97%, respectively which indicated the high accuracy of the results. Also the results of factor analysis showed that the role of land use in sedimentation of the study area is more than other factors.

The comparisons between observed and simulated sediment loads showed that the model has fairly acceptable accuracy for Badavar catchment. A more accurate estimation of erosion and sediment yield can be made by providing accurate data and the best management practice is highly recommended for the dam sustainability, because of the proximity of Badavar cahchment erosion to the Karkheh dam.
Key word: Erosion, Sediment, Factor Analysis, SWAT, Badavar.